© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 32-TRANSCODER-HP
    7. 43315
    Prev
    Next
    INDEX
    Explanations

    child

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     parents
    -0.28
    å¥Ĥ
    -0.28
    çıŃç»Ħ
    -0.25
     conveyed
    -0.25
    ÑģÑĤÑĢа
    -0.24
     bone
    -0.24
     rolled
    -0.24
     bod
    -0.24
    dden
    -0.24
     accessed
    -0.24
    POSITIVE LOGITS
    æĶ¯æĮģ
    0.41
    support
    0.33
     supports
    0.33
     support
    0.32
     Support
    0.30
     CIT
    0.29
     supporting
    0.29
    upport
    0.29
     Supports
    0.28
    اع
    0.28
    Activations Density 0.001%

    No Known Activations